Helper functions to compute GO enrichment tests using the Bioconductor R-packages GOstats and GSEABase

# Source this R script to install goEnrichment and all dependencies:
if (!require(devtools)) {
  install.packages("devtools")
  require(devtools)
}
packages <- c("GOstats", "GSEABase")
if (length(setdiff(packages, rownames(installed.packages()))) > 0) {
  source("http://bioconductor.org/biocLite.R")
  biocLite(setdiff(packages, rownames(installed.packages())))
}
install_github("asishallab/goEnrichment")

Copy and edit the supplied example input file ./inst/input.R to your needs and run it:

Rscript path/2/goEnrichment-package/exec/runGoEnrichment.R your_edited_input.R

If you want to test enrichment with non Arabidopsis or C. hirsuta genes you need to provide a custom background GeneSetCollection. In an interactive R shell run the following. Note, that you need a Gene Ontology (GO) annotation table of your background. The table is required to have three columns:

1. V1 must hold the GO term accessions,
2. V2 the evidence codes, e.g. IEA, and
3. V3 must hold the gene identifiers (accessions).

Having such a table run the following code:

require(GOstats)
require(GSEABase)
goa.tbl <- read.table( "path/2/go_annos_table.txt", stringsAsFactors=FALSE, sep="\t" )
univ.go.annos <- sort(unique(goa.tbl$V3))
goFrame <- GOFrame(univ.go.annos, organism = "Homo sapiens")
goAllFrame <- GOAllFrame(goFrame)
gsc <- GeneSetCollection(goAllFrame, setType = GOCollection())
# Now save both your GeneSetCollection and the GO annotation table in binary format:
save(gsc, file="custom_gsc.RData")
save(goa.tbl, file="custom_goa.RData")

The above two prepared custom obejcts can subsequently be used in an adjusted input.R as explained there.

If you want to test goEnrichment you can run it with test_input.R. On a *nix like system that would be:

Rscript path/2/goEnrichment/exec/runGoEnrichment.R path/2/goEnrichment/test_input.R